PERIODISMO DIGITAL
2. Contexto digital y educación:
The findings of this study show two status reporting issues. First, there was distortion of status reporting to the ECPR panel in the case of project Gamma. Second, that the Deaf Effect occurred in the case of project Gamma while it did not occur in the cases of project Alpha and Beta.
The findings of this study also show that the existing SISO status reporting process is dependent upon the voluntary disclosure of information by the projects. Projects Beta and Gamma encountered similar problems. How they responded to the problems determined the success or failure of the process. Project Beta responded by disclosure of the problems to the panel, which led to their resolution. Project Gamma responded by a systematic reduction in communication about the problems, which resulted in the ultimate failure of the project. Despite the monitoring by the IV&V, the SISO process and a steering committee, the project Gamma management was able to avoid disclosure of the issues until the first attempt at implementation. This illustrates that the IV&V, which acquired its information largely through attendance at various project management meetings, was dependent upon the project team in order to get information to assess the status of projects.
Table 5.3 illustrates the differences in the information gathering opportunities between projects Alpha and Beta and project Gamma. The Agency Commissioner in projects Alpha and Beta was highly involved in the monitoring process as part of the steering committee, as a participant in the ECPR process. They also received status reports directly from the IV&V vendor. In project Gamma, however, the Agency Commissioner was largely left out of the process. SISO was involved in the process as part of the steering committee and by virtue of their review meeting prior to the ECPR meeting in projects Alpha and Beta. In project Gamma however, GTA was not involved in the process. Agency executives and stakeholders populated the steering committee in projects Alpha and Beta whereas that of project Gamma was comprised of relatively low- level stakeholders.
Table 5.3: Differences Between the Projects
Project Alpha Project Beta Project Gamma
Involvement of Commissioner Part of Dashboard Review Process Part of Dashboard Review Process Oblivious (Interviewee Harvey, CIO, County 1 School District) GTA Review Meeting
Part of Process Part of Process Not part of process
Steering Committee Formal, Populated with top department and SISO personnel
Formal, Populated with top department and SISO personnel
Informal. Populated with low-level department personnel (Interviewee Harvey, CIO, County 1 School District). No SISO
personnel (Interviewee Sherman, SISO ECPR Facilitator). IV&V Participation in Project Meetings
Yes Yes Yes
Full Disclosure of Issues in Project Meetings
Yes Yes No (Interviewee
Braxton, Contractor Project Manager) IV&V Access to
Commissioner
Yes Yes No (Interviewee
Joaquin, IV&V Project Gamma) EPD Agency
Reporting structure
To CIO To CIO Acted as CIO,
reported to Commissioner State Government Relationship Reports Directly to Governor Reports Directly to Governor Reports to Constitutional Officer Results In process. No issues noted. Finalizing. Major issue occurred in 7/06 but overcome. Finalized in 12/05 but determined to have failed in 10/06.
5.5.1 Distortion of the Message
The distortion of the status report in project Gamma occurred in the reporting of status of project Gamma as “green” during the development and pilot stages of the project, when by testimony of the project management (Braxton, Contractor Project Manager and Jackson, Project Gamma Technical Project Manager), the status was actually “yellow” or “red”. The responsibility for the distortion was placed on Forrest. Table 5.4 displays the various theories that were discussed in section 5.2.2 that might cause distortion of the message and whether they occurred in project Gamma.
Table 5.4 – Distortion Causing Conditions in Project Gamma Source of
Distortion
Theory / Source
Condition in Project Gamma
Errors of Understanding
Status
Reporting Bias (Snow and Keil, 2002)
This condition does not seem to exist. Braxton, Jackson, Forrest agreed on the facts. They differed on the required action.
Bias in Reporting
Status
Reporting Bias (Snow and Keil, 2002)
Upper Echelons
In the history of project Gamma, we see evidence of self-efficacy, a political interpretation of facts and a control fixation.
(Hambrick and Mason 1984) Reaction to Management Responses to Reports of Bad News Critical Upward Communication (Tourish and Robson (2006),Morrison and Milliken (2000))
Forrest had not experienced any bad repercussions from reporting bad news on this project. The actual responses by the ECPR and SISO were very encouraging as reported by project Beta. However, Forrest did not know this.
Intrusive Supervision
Athanassiades (1973)
No evidence exists that show that Forrest considered the supervision intrusive. Adversarial Approach by Regulators Regulatory Theory (Scholz 1984, 1991)
No evidence exists that an adversarial relationship existed between IV&V, SISO, or ECPR and Forrest.
The table shows that the theories that had been considered in section 5.2.2 do not explain the occurrence of distortion in the status report of project Gamma with the exception of Snow and Keil’s status reporting theory. Status reporting distortion may occur as a result of misunderstanding of the situation (Snow and Keil 2002; Snow, et al. 2007). This situation did not occur in project Gamma. Jackson and Braxton indicated that there were serious problems. Forrest did not disagree with the facts, but rather with the seriousness and the proposed solution. Similarly, the responses of supervisors and regulators may cause distortion (Athanassiades 1973; Morrison and Milliken 2000; Scholz 1984; Scholz 1991; Tourish and Robson 2006). In project Gamma, we do not see any adverse responses to bad news by management, IV&V or the ECPR panel to
Forrest that would cause such distortion. However, the findings do show potential biases to reporting in that Forrest considered the situation in which he found himself a highly political situation in which he felt he had to control the communications of the project so that the project would be viewed correctly. Thus he would bias the reporting in an effort to avoid loss of power.
There is however, another theory which could add additional insight into this situation: agency theory (Eisenhardt 1989; Jensen and Meckling 1976). Agency theory seeks to address the situation in which one party (the principal) delegates work to a second party (the agent) under a contractual relationship. Two problems arise under the contract: the
agency problem where the goals of the principal and agent differ and it is difficult and/or expensive for the principal to verify that the agent is fulfilling his desires. This manifests
itself in two issues: moral hazard in which the agent does not perform as agreed under
the contract (shirking) and the principal does not have the information or the ability to
acquire the information to verify the performance of the agent; and adverse selection in
which the agent misrepresents his abilities in order to get the job and the principal cannot verify his statements. The second issue in the agency relationship is risk sharing in which the principal and agent have different attitudes toward risk. The object of the theory is to identify the most effective contracting mechanism between them; whether behavior based or outcome based mechanisms are more effective (Eisenhardt 1989).
Two studies have applied agency theory in escalation situations. Harrison and Harrell (1993) investigated adverse selection. They found that managers who initiated
unprofitable projects tended to consider them. This is due to the information asymmetry that existed which allow them to hide their incompetence. Where adverse selection did not exist, the agents behaved in the interests of the principal. Tuttle, Harrell and Harrison (1997) investigated moral hazard conditions in IS implementation situations. They found that agency theory predicted the behavior of students in moral hazard situations well. In laboratory experiments students indicate willingness to implement systems with known quality issues under conditions of moral hazard. This result was tempered with the finding that this effect is moderated by the individual’s moral code. In some cases, it overrides self-interest to not implement the system.
In the situation of project Gamma, a definite moral hazard situation existed. Moral hazard exists when an agent has an incentive to not perform as expected by the principal and information asymmetries exists so that the principal cannot verify the performance of the agent (Eisenhardt 1989; Tuttle, et al. 1997). In that project, as discussed above, Forrest viewed his position as politically charged and that the maintenance of his position required the perception that all was well with the project. There was therefore a significant temptation to shirk. Similarly, table 5.5.1 shows that significant information asymmetries existed. Forrest controlled the communications external to the project so that external entities only received the information he wanted to provide them. Project Gamma therefore conforms to an agency theoretic view of project reporting.
Thus in this situation, I suggest that Forrest’s perception of a highly politicized
environment and need to control the information flow created an incentive to distort the status of the project which when combined with the information asymmetries in the project created a moral hazard situation resulting in distortion of the project status in the reports to the ECPR and the Agency Commissioners.
5.5.2 Evaluating Project Gamma using the Individual Model from this
Dissertation.
The findings of this study showed that the Deaf Effect occurred at least three times in the case of project Gamma. This section addresses how well the model developed in
chapters two through four explains the occurrence of the Deaf Effect in project Gamma and then addresses additional factors that could be added to the model that are
illustrated in the case study.
The individual model developed in the previous chapters placed the subject in a
particular organizational environment. For the model to be validly applied, it must first be ascertained that that environment existed in this case.
Four factors were included in the environment used in the scenarios:
1) External pressures that moved the decision maker to continue the current course of action;
2) A decision maker who could not perform the evaluation on his/her own;
3) A report of bad news that did not communicate the potential disaster situation in terms that unambiguously indicated a coming disaster;
4) Severe time constraints on the length of time to evaluate the situation and make a decision that prevented either evaluation of the situation or cut off alternative solutions.
There is no direct evidence of Forrest’s superiors constantly pressuring him to deliver. However, it is obvious that he felt some pressure. He stated that he had been brought in “to make it happen” whereas it had failed twice before. When presented with issues by Jackson, he repeatedly stated that it was “politically impossible” to take any alternative course. On the second point, Forrest did not have the technical skills to evaluate the technical performance of the project. He was reliant upon the representations of others. Third, while in retrospect, the warnings of the October, 2005 memo seem very prescient, at the time, Forrest may have been able to rationalize continuing by considering them to be overblown or perhaps he considered them capable of resolution by other means. Lastly, he definitely felt pressure to deliver a system by 12/05, which prevented him from considering any other alternatives at that late date.
Given that the same conditions existed in the case that existed in the experiments, I now consider whether the model developed in chapters two, three and four holds in this case. Since there was an occurrence of the Deaf Effect, based on the model, we would expect that 1) the message of bad news was not considered relevant by the decision maker; 2) that there would be a perception of little risk in the project; 3) there would be a
perception of low credibility in the bad news reporter; and 4) the bad news reporter would be not role prescribed to report bad news.
Considering point 1, in the findings, there are statements that Forrest considered Jackson’s concerns to be irrelevant because they were 1) politically not feasible and 2) not based in knowledge of everything going on. He stated that a system had to be technically, operationally, economically and politically feasible and political feasibility would trump all other concerns. This meant that in order to maintain his political position, he couldn’t back track on the course of action. Therefore, given the situation, the report of bad news was not as relevant to the decision as the political factors, so, we see here a low Relevance ascribed to the report of bad news.
Considering point 2 above, in the findings we see no indications that Forrest perceived the risk in the project to be exceptionally high or at least not beyond his capability to manage it. He believed that Jackson did not know about his work with the legacy systems that were going to resolve the data loading problem and therefore Jackson’s concerns did not escalate his perception of risk. He also perceived the legacy data collection system as his contingency plan in the event of failure of the project data loading system. Therefore, given that he perceived he had multiple paths to accomplish his goal, he did not perceive that there was a high risk of project failure..
Considering point 3, the credibility of the bad news reporter, we understand from source credibility literature that Credibility is formed from two primary constructs, trustworthiness and expertise (Pornpitakpan 2004). We see that Forrest held the reporter, Jackson in high regard. He considered him an excellent project manager. But he felt that his knowledge of what was going on in the rest of the project was lacking. He therefore lacked expertise. Therefore, his Credibility was low in this area.
For point, 4, Role Prescription of the bad news reporter, we see that there was no specific aspect of his job which required Jackson, who reported the bad news in this project, to “blow the whistle” on the project. However, it is usually assumed that a project manager will report serious project problems to his superiors.
Considering these factors within the model developed above, we see the following conditions (figure 5.4). Forrest considered the risk of the project as low which would have the effect of reducing his willingness to change course of action. Because Forrest considered Jackson to lack complete knowledge of the situation, he had a lowered opinion of his credibility as a bad news reporter. This low Credibility acted to lower the perceived Decision of the report of bad news. While Role Prescription of the bad news reporter is seen as high, given the combination of low Credibility with high Role
Prescription, we have seen that the effect size of Credibility is greater than that of Role
Prescription (chapter 3) resulting in a low Relevance of the message. The low Credibility would also act to lower the perceived risk of the project. Given this set of conditions, a low relevance and low Risk Perception, the model predicts that Forrest have continued the current course of action which is what I found.
Thus the individual model holds in predicting the decision to be made by the decision maker in this situation.
5.5.3 Examining Other Escalation Factors
In this section, I seek to expand the model to identify other factors by examining the other escalation determinants as defined in Staw’s (1997) classification of escalation. I examine each one in turn
5.5.3.1.
Illusion of Control
One of the determinants for project escalation referenced by Staw (1997) is that of optimism and illusion of control. Keil, Depledge and Rai (2007) found that when a subject perceived that they had the ability to market around software quality issues they were more likely to not recognize the software quality problems and to escalate the commitment to the project by shipping a defective software product.
In the current case, the project manager was supremely confident in his knowledge and abilities and it is therefore possible that he might consider himself able to deal with a software product with issues on implementation. In terms of the model, it can be hypothesized that when a project manager perceives that he has substantial skills and
Role Prescription